Microsoft didn't even have anyone working on IE for most of the five years between IE6 and IE7 (2001 - 2006). Without competitors like Firefox to keep web standards and cross-browser compatibility alive through that era, it would have taken a lot longer to dig ourselves out of the IE6 era.
(Remember when IE had 90% market share, and web developers routinely wrote IE-only code?)
But IE7 wasn't much of an upgrade over IE6. Neither was IE8. Browsers didn't take the next step until Google got fed up with javascript rendering speeds and took it into their own hands; now every browser is fast.
Firefox AND Opera made the web bearable in the years between IE6 and Chrome, but they hardly are responsible for the arms race that is taking place now.
IE8 is a huge upgrade over IE6. For one thing, at ship time it implemented the entirety of the then-current CSS 2.1 draft Just comparing it to IE7, it added generated content, automatic counters and quotes, tons of fixes to floats, outlines, lots of box model bugfixes. the list goes on and on. It's not a huge upgrade in terms of _Javascript_, but in terms of CSS it's light and day when you compare to IE6. Oh, and IE8 moved to a process-per-tab model.
The thing I think you're missing in the evolution of IE after the team was disbanded post-IE6 is that it takes time to build up a project team from 0 to hundreds of people, especially when the codebase is preexisting and large. IE7 was them starting to put a team together again, doing catchup on obvious UI features like tabbed browsing. IE8 was them finally having enough people to execute well on some things (CSS 2.1, process-per-tab) and try to play catchup on other things (JavaScript performance). IE9 is them having a team big enough to work on all parts of the browser at once.
You also have your history slightly wrong on the JS perf. Chrome was first announced on September 1, 2008; before that some people knew that Google was working on a browser, but the details were secret. The WebKit project initially landed the SquirrelFish Extreme jit on September 18, 2008; it's pretty clear that this was in the works before Chrome's existence was announced. Mozilla landed the initial Tracemonkey implementation in their main development tree in mid-August 2008 (weeks before Chrome was announced), and had been working on it for some time before that. So all three (Firefox, Safari, Chrome) were working on JITs in parallel before Chrome was announced; it's not like Chrome shipped a JIT and everyone else suddenly decided they had to do it as well. Had Chrome never existed, Firefox and Safari (and Opera, which joined in on the game in early 2009) would still have competed for faster javascript; IE9 would still have ended up doing a jit, or risking being left in the dust.
Oh, and the Sunspider benchmark, for example, was first released in December 2007, when both WebKit and Mozilla were actively working on their interpreters' performance and starting to work on the JITs they'd have by September 2008. Again, the competition for faster JavaScript was very much in play already.
Now Chrome has contributed to the competition, as have all the other players. But they sure didn't start it.
"The WebKit project initially landed the SquirrelFish Extreme jit on September 18, 2008; it's pretty clear that this was in the works before Chrome's existence was announced."
In fact, the Apple WebKit folks have said that they began the Squirrelfish Extreme project the day after Tracemonkey was announced. It was quite an engineering accomplishment to have an initial stab at a JIT in a month and have it shipped to users in a few more, especially considering it took Mozilla about a year (Summer '08 - June '08) to ship theirs. And everyone knew about Tamarin's development over the previous couple years - so it ended up being a really amazing coincidence that all three JITs were announced within a one month period.
Remember when mozilla blew so badly that IE6 was the best browser around? Firefox wasn't even around for most of the 2001-2006 era, until somebody finally decided the mozilla all in wonder mega suite was a crime against nature.
Sure, firefox keeps IE in line, but mozilla also has a history of jumping the tracks without strong competition.
(Remember when IE had 90% market share, and web developers routinely wrote IE-only code?)